J. Am. Chem. SOC.1992,114, 8328-8329
8328
A Regiospecifically Oxygen-18 Labeled 1,2,4-Trioxane: A Simple Chemical Model System To Probe the Mechanism(s) for the Antimalarial Activity of Artemisinin (Qinghaosu) Gary H. Posner* and Chang Ho Oh Department of Chemistry The Johns Hopkins University Baltimore, Maryland 21 21 8 Received July 2, 1992
Among infectious diseases, malaria is the most widespread. Worldwide, about 300 million people are infected, and 1-2 million die each year.' As malarial parasite resistance to alkaloids such as quinine has increased, new non-alkaloidal antimalarial drugs such as the 1,2,4-trioxaneartemisinin (qinghaosu, 1) have become increasingly important.2 Current understanding invokes hemin-catalyzed reduction of the trioxane unit in artemisinin as the key step activating it into one or more cytotoxic compounds that kill malarial parasite^.^ The hemin-rich internal environment of malarial parasites is thought to be responsible for the selective toxicity of trioxanes like artemisinin toward these para~ites.~To gain a better understanding at the molecular level of this antimalarial activity of activated artemisinin, we report here the preparation and cleavage reactions of simpler, regiospecifically oxygen-18 labeled 1,2,4-trioxane Sc that also is a potent antimalarial c o m p o ~ n d . ~ Oxygen-18 was introduced from 180Hzduring the conversion of nitrile 2 into methyl ketone 3 (Scheme I).596 1,2-Dioxetane formation, using either singlet molecular oxygen or triethylsilyl hydr~trioxide,~ gave intermediate dioxetane 4 that rearranged under the influence of (tert-butyldimethylsilyl) triflate into 1,2,4trioxaneSa? Fluorideinduced desilylation produced trioxane alcohol Sb. The location of I8O at position 4 in the 50% lEOenriched 1,2,4-trioxane Sb was established by I3C NMR spectroscopy, in comparison with an unlabeled version of trioxane Sb, showing doublets at 105.3 and 100.2 ppm characteristic of carbon atoms at positions 3 and 5 (Scheme I);6 only carbons 3 and 5 showed onebond I8O-inducedisotope shifts. This isotopic labeling experiment supports the proposed zwitterionic peroxide pathway (see arrows in structure 4) for rearrangement of keto dioxetanes such as 4 into 1,2,4-trioxanessuch as Sa.s Trioxane alcohol Sb (1) UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases: Tropical Diseases Progress in Research, 1984-1 990; World Health Organization: Geneva, Switzerland, 1991; pp 29-40. See also: Maegrith, B. Clinical Tropical Diseases, 9th ed.; Blackwell Scientific Publications: Oxford, England, 1989; pp 201-246. (2) For reviews, see: (a) Klayman, D. L. Science 1985, 228, 1049-1055. (b) Luo, X.D.; Shen, C. C. Med. Res. Rev. 1987, 7, 29-52. (3) (a) Zhang, F.;Gosser, D. K., Jr.; Meshnick, S.R. Biochem. Pharm. 1992,42,238-242. (b) Vennerstrom, J. L.; Eaton, J. W. J . Med. Chem. 1988, 31, 1269-1277. (4) Scott, M. D.; Meshnick, S.R.; Williams, R. A.; Chiu, D. T.-Y.; Pan, H. C.; Lubin, B. H.; Kuypers, F. A. J . Lub. Clin. Med. 1989, 114, 401-406. (5) Posner, G. H.; Oh, C. H.; Gerena, L.; Milhous, W. K.J . Med. Chem. 1992, 35, 2459-2467. (6) (a) Diakur, J.; Nakashima, T. T.; Vederas, J. C. Can. J . Chem. 1990, 50, 1311-1315. (b) For a review, see: Vederas, J. C. Nor. Prod. Rep. 1987, 4, 277-337 and references therein. (7) Posner, G.H.; Oh, C. H.; Milhous, W. K. Tetrahedron Lett. 1991,32, 4235-4238 and references therein. (8) (a) Jefford, C. W.; Boukouvalas, J.; Kohmoto, S. Helu. Chim. Acta 1983, 66, 2615-2620. (b) Jefford, C. W.; Boukouvalas, J.; Kohmoto, S. J . Chem. Soc., Chem. Commun. 1984,523-524. (c) Jefford, C. W.; Jaggi, D.; Kohmoto, S.; Boukouvalas, J.; Bernardinelli, G.Helu. Chim. Acta 1984, 67, 2254-2260. (d) Jefford, C. W.; Boukouvalas, J.; Kohmoto, S.; Bernardinelli, G.Tetrahedron 1985,41,2081-2088. (e) Jefford, C. W.; Favarger, F.; Ferro, S.; Chambaz, D.; Bringhen, A.; Bernardinelli, G.; Boukouvalas, J. Helu. Chim. Acta 1986.69, 1778-1786. (0 Jefford, C. W.; McGoran, E. C.; Boukouvalas, J.; Richardson, G.; Robinson, B. L.; Peters, W. Helu. Chim. Acta 1988, 7 1 , 1805-1812. (g) Jefford, C. W.; Verlade, J.; Bernardinelli, G.Tetrahedron Lett. 1989, 30,4485-4488.
was treated with tosyl chloride and triethylamineto form trioxane tosylate S C . ~Tosylate Sc was chosen for reduction experiments because it is an active antimalarial agentSand because oxyanionic intermediatesformed during the reduction process were expected to be trapped intramolecularly by displacement of the tosylate group. Trioxane tosylate Sc was treated separately with several established reducing agents. Samarium diiodide (at 25 "C)? zinc (at 25 OC),l0 Ph3CLi (at -78 oC),lla and Bu3SnH/AIBN (in refluxing benzene)l2all produced 1,Edioxolane tosylate 8 within 4 h in good to excellent yields. "C NMR spectroscopy confirmed that l8O was located exclusively at the exocyclic hemiacetal position! No reaction occurred when trioxane Sc was treated with typical peroxide-cleaving reagents such as M%S,Ph3P,or sodium dithionite.lib The mechanism in Scheme I1 is proposed to account for the generation of dioxolane 8." Electron transfer to trioxane Sc gives a radical anion; specifically, radical anion 6 unzips with extrusion of methoxide anion to form trioxane ring-cleaved intermediate 7 that rezips to produce the observed dioxolane 8. Formation of a 1,3-dioxolane by the deoxygenation of a 1,2,4trioxane has physiological relevance because malarial parasites operate on trioxanes like artemisinin to produce the corresponding diox01ane.I~ Had radical anion 6 been formed with the opposite arrangement of radical and anionic centers, then methoxy radical extrusion and rezipping would have led ultimately to intramolecular oxyanion displacement of the pendant tosylate group to form cyclic ether 9,Is Indeed, cyclic ether 9 was produced in excellent yield when hydroxy tosylate 8 was treated with the base sodium hydride and also when trioxane Sc was treated at 0-25 OC with dimethylcopperlithium,a basic reducing agent.I6 Also, treating the trioxane iodide analogous to trioxane tosylate Sc with dimethylcopperlithium gave cyclic ether 9 directly and exclu~ive1y.l~ Trioxane tosylate Sc was treated at room temperature in THF also with two different sources of ferrous ions to simulate the biologically important erythrocyte hemin-catalyzed cleavage of the trioxane unit in artemi~inin.~ Although ferrous bromide and hemin/PhCH3H (no reaction occurred without PhCH$H)3 both induced rupture of the peroxide linkage in trioxane Sc, products 11, 12, and 14 (different from dioxolane 8) were produced." Ring-contracted tetrahydrofuran acetal 11, aldehyde 12, and hydroxy dioxolane 14 are typical of the metabolites formed from trioxanes such as artemisinin in the presence of rat liver microsome~;'~ i n d d , when we treated artemisinin with ferrous bromide, an acetal like 11 and a hydroxy dioxolane like 14 were formed along with a dioxolane like 8. When hemin/PhCH2SH was used on trioxane Sc, PhCHzSSCH2Phwas formed, analogous to the protein thiol oxidation products (Le., disulfides) observed when artemisinin itself and hemin interact in the presence of red cell membranes.17 The radical mechanism in Scheme 11, similar to the mechanism proposed earlier for the very high temperature (9) (a) Girard, P.; Namy, J. L.; Kagan, H. B. J . Am. Chem. Soc. 1980, 102, 2693-2698. (b) Molander, G. A.; La Belle, B. E.; Hahn, G.J . Org. Chem. 1986, 51, 5259-5264. (c) Hanessian, S.; Girard, C.; Chiara, J. L. Tetrahedron Lett. 1992, 33, 573-576. (d) For a review, see: Soderquist, J. A. Aldrichimica Acta 1991, 24, 15-23. (10) Jefford, C. W.; Rosier, J.-C.; Boukouvalas, J. Heterocycles 1989,28, 673-676. (11) (a) Adam, W.; Heil, M. Chem. Ber. 1992, 125, 235-241; J . Am. Chem. Soc. 1992,114,5591-5598 and references therein. (b) Idowu, 0. R.; Maggs, J. L.; Ward, S.A.; Edwards, G. Tetrahedron 1990,46, 1871-1884. (12) For a review, see: Neumann, W. P. Synthesis 1987, 665-683. (13) All new compounds were characterized spectroscopically and by combustion analysis and/or high-resolution mass spectrometry. (14) (a) Baker, J. K.; Yarber, R. H.; Hufford, C. D.; Lee, IPS.;EISohly, H. N.; McChesney, J. D. Biomed. Enuiron. Mass Spectrom. 1988, 18, 337-351. (b) Leskovac, V.; Theoarides, A. D. Comp. Biochem. Physiol. 1991, 99c, 391-396. See also: Wei, Z.; Pan, J.; Li, Y. Planta Med. 1992,58,300. (15) For a structurally related cyclic ether, see: Jung, M.; Li, X.;Bustos, D. A.; ElSohly, H, N.; McChesney, J. D. Tetrahedron Lett. 1989, 30, 5973-5976. (16) (a) Posner, G. H.; Ting, J.-S. Synth. Commun. 1973,3, 281-285. (b) Posner, G.H.; Sterling, J. J. J. Am. Chem. SOC.1973, 95, 3076-3077. (17) Meshnick, S.R.; Thomas, A.; Ranz, A.; Xu, C.-M.; Pan, H.-Z. Mol. Biochem. Parasitol. 1991, 49, 189-190.
0002-1863/92/ 1514-8328303.00/0 0 1992 American Chemical Society
8329
J . Am. Chem. SOC.1992, 114, 8329-8331 Scheme I
q)
Scheme II
H6
I
OTS
5C
bTs
1
L
OTs
9
J
(190 "C) pyrolysis of artemisinin,Is is offered to account for these room-temperature results. Iron(I1)-induced cleavage of the peroxide bond in trioxane 5c ieads to radical intermediates 10a and lob in about a 2:l ratio: C-C bond cleavage of 1Oa initially produces labile ring-contracted tetrahydrofuran acetal 11 (characterized by IH and I3C NMR) with I8O located in the acetoxy group as shown in Scheme I1 (mass spectrum, M CH3C01*0)and then produces stable electrophilic tetrahydrofuran aldehyde 12 lacking lsO. 1,5-Hydrogen atom abstraction in radical intermediate 10b ultimately leads to stable dioxolane alcohol 14 as a mixture of two diastereomers with I8O not located in the methoxyl group (mass spectrum M - CH30). Subsequent oxidation of this isomeric mixture of alcohols 14 gave the corresponding dioxolane ketone IS as a single product.13 The overall yields of isolated aldehyde 12 and hydroxy dioxolane 14 ranged from 60 to 70%. In summary, these reactions of trioxane Sc for the first time (1) provide firm mechanistic evidence that deoxygenation of a 1,2,4trioxaneinto the corresponding 1,3-dioxolane occurs via a tandem unzipping-zipping process and (2) show that trioxane cleavage by fer" iow follows a Merent mechanistic course and leads to different product8 than Moxane cleavage by nonferrous redwing agents. These results may help the development of better antimalarial t r i o x a n e ~ . ~ ~ * * ~
A.tWe thank the Environmental Health Sciences Center, School of Hygiene and Public Health, The Johns Hopkins University, and Dr. Henry Sonneborn for financial support, Dr. Judith Stamberg (Union Memorial Hospital, Baltimore) for suggesting the use of a thiol to activate hemin reduction of trioxane (18) Lin, A. J.; Klayman, D. L.; Hoch, J. M.; Silverton, J. V.; George, C. F. J . Org. Chem. 1905, 50, 4504-4508. (19) For a recent review of the chemistry and biological activity of artemisinin and related antimalarials, see: Zaman, S. S.; Sharma, R. P. Heterocycles 1991, 32, 1593-1638. (20) For recent use of '*O labeling to study the mechanism of conversion of dihydroartemisinicacid into artemisinin, see: Acton, N.; Roth, R. J. J . Org. Chem. 1992,57,3610-3614. We thank Dr. Acton for a preprint of this article.
Sc, Professor Craig Townsend (Johns Hopkins) for a helpful discussion about labeling, and Professor Steven Meshnick (City University of New York Medical School) for some helpful preprints and discussions about porphyrin adducts with trioxanes like artemisinin. Supplementary Material Available: Listing of full experimental details and spectral data for compounds 3, Sa-, 8 , 9 , 11, 12, 14, and 15 (36 pages). Ordering information is given on any current masthead page.
Remarkable Regioselectivity in the Chemical Glycosylation of Clycal Acceptors: A Concise Solution to the Synthesis of Sialyl-Lewis X Clycal Samuel J. Danishefsky,* Jacquelyn Gervay,la John M. Peterson,lb Frank E. McDonald,l' Koshi Koseki,Id Takeshi Oriyama,lc and David A. Griffith" Department of Chemistry, Yale University New Haven, Connecticut 0651 1
Chi-Huey Wong and David P. Dumas Department of Chemistry, The Scripps Research Institute, 10666 North Torrey Pines Road La Jolla. California 92037 Received May 21, 1992
The cell-surface-bound polysaccharide sialyl-Lewis X antigen (SLe", has recently been identified as a ligand for binding to the cell-adhesion molecules ELAM-1 and CD-62.3 These proteins are expressed on cell membranes in response to tissue injury, and
0002-7863/92/ 15 14-8329%03.00/0 0 1992 American Chemical Society
